Fortinet, Inc. (FTNT) Earnings Call Transcript & Summary

Thanks, everyone, for joining us again. My name is Sterling Auty. I'm a software analyst here at JPMorgan. Very happy to have with us the team from Fortinet. We have Ken Xie, who is Founder and CEO; and Keith Jensen, who's CFO. I am going to turn the microphone over to Keith just to give a couple of safe harbor remarks and then we're going to jump right into it. Keith?

Thank you, Sterling. I'd like to remind everyone that we may make forward-looking statements during today's fireside chat. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those projected in these statements. Please refer to our SEC filings, in particular the risk factors in our most recent Form 10-K and Form 10-Q and to other reports that we may file from time to time with the SEC for additional information on factors that may cause actual results to differ materially from our current expectations. All forward-looking statements reflect our opinions only as of the date of this presentation, and we undertake no obligation and specifically disclaim any obligation to update forward-looking statements.

So I didn't catch all that. Could you repeat that, Keith?

Yes, I'll go slow.

Ken, I think there's a perception from some investors that you got this big tailwind because of COVID. Can you walk through what were the benefits that you saw and maybe some of the headwinds that you saw through the pandemic. You got to unmute. I did that earlier, so don't worry about it.

Yes. When the pandemic start, we do see a little bit rush by end of Q1. And then Q2, we mentioned there's some product like a FortiToken authenticator has some kind of a quick ramp-up. So I think we gave the number, Keith, if I confirm, it's about $10 million -- $5 million to $10 million range. And then after that, it's pretty, pretty normal. I don't see any benefit or anything from the pandemic. But I have to say, starting from this year, once this thing steady and we open up, we do see going forward in the next 1 to 2 years, there will be a very healthy spending environment and whether the new infrastructure like SD-WAN 5G or kind of working with service provider to offer SASE or kind of work from home, the new infrastructure to supporting work from home, all these things we do see pretty healthy growth going forward.

That's great. And then just wanted to shift talk. SD-WAN is an area that you shifted the company to focus on very early and you definitely benefited from it. Can you give us a sense how much of the business today is being driven by SD-WAN?

Last quarter, it's about 14% come from SD-WAN, but we do see the growth almost double, I think probably more than double year-over-year. The market itself grow 50% year-over-year. We grow 100%. But it's also we see -- the market probably keeping growing about 40%, 50% in the next 3 to 5 years. And we do believe we'll keep on gaining market share because some other competitor much more slowed down quite a lot. Some of them no longer in the Gartner Magic Quadrant as a leader, some of them because all come from acquisition, lot of people leaving and were behind on the product development, but we keep up all this development innovation and also leverage our ASIC advantage, cost advantage. So we do see so far keeping growing like 100% year-over-year and also very healthy going forward.

So for those that are unaware or have trouble understanding the difference, Zscaler just reported, they just showed 60% growth accelerating again. Sometimes the messaging can get a little bit confused. Where do you play in the market versus what a solution like Zscaler would offer?

Our SD-WAN and also SASE, we integrate into the FortiOS and also same as a zero trust and the same like the 5G. And also we're more working with the service provider carrier like AT&T, BT, all these, to supporting and offered SD-WAN. So that's part maybe take a little bit long term. But we do see they have a much better position to offer the SASE to their customer base and leverage their own infrastructure. Even they are more a little bit slow right now, but there's a huge potential going forward. But also, we feel the tight integration among the same FortiOS is very, very important, not just for the customer but also for the service provider. And that's all for better SASE, better security and also better Zero Trust Network Access together with SD-WAN 5G. As Keith probably also see, quite some interest pipeline and maybe he can add some more.

No, I think the SD-WAN part of the business is no great secret. We did, what, 14% of the business in billings last quarter, Ken, as you mentioned. I think the other part of it is the size of the deals. $15 million of our $66 million deals last quarter were SD-WAN transactions, and that's about 23% of our $1 million deals. And yes, when we look at the pipeline, it's something that we continue to look at each week in terms of what our expectations are for continued growth as we move through the second quarter and the rest of the year, I think we feel very good, as Ken mentioned before, about comparing ourselves to market growth rates and wanting to continue to take market share.

Absolutely. And again, just to help the investors understand, when you say that you sell SD-WAN, what are you actually selling? Is it a service? Is it an appliance? What is this solution?

I think first, the product need to be there. So that's where we see the product revenue keeping growing more healthily. And probably in the space, we are the only one keeping -- like last quarter, the product revenue growth, 25%, much higher than any other competitors. Some has low single digits, some has no growth. That's also will enable the future service revenue growth and especially our own product. We offer much more function and enable much more service. Some of them, we just left money on the table, will be future potential, the additional service growth there. We treat that as kind of investment for the future service revenue. But SD-WAN, we see is a lot of -- not just enterprise customers, but also work from home, other retail and even among enterprise and routing traffic between data center, they all need this new SD-WAN technology, which we leverage ASIC and also integrate with other function is doing quite well.

You alluded to your peers in terms of the growth. And Palo Alto showed good growth in what they call Prisma Access. But is that really comparable to what you're providing? We always think about them as your traditional competitor for on-premise networking security. What is the comparison and the impact on your business from Palo Alto for SD-WAN?

We are more putting this FortiGate -- FortiOS and also integrate with like the SASE, some other Zero Trust Network Access. Some part has overlap, some other part has not because we view that securing whole infrastructure is very, very important. And some competing with their network sites, some competing with their kind of the SASE side. But for us, it's a single product, single OS. That's also the reason we are the only vendor as the single same product FortiGate in both Gartner Leaders Magic Quadrant for network security and SD-WAN, the same OS in product, right? So you don't see any other competitor have this one. That's how powerful the integration, how powerful the computing power we can offer using this new platform, which also working well with the other product, we call the Fabric, which is [indiscernible] product, most of the company internal development as they integrate, automate from day 1 as different than competitor come from acquisition, which is also very profit for us, but also grow like 50% year-over-year last quarter. We do see a lot of potential from that angle also.

You mentioned SASE a couple of times as a term that's, I think, frequently being used across the market. And I think different vendors use it differently. So for you, what are the solutions that you're offering for SASE?

Like I said, we do want to working with service provider because that's their business where they offer their service. SASE, I see that the #1 benefit is really leverage service and kind of offload some of IT burden there, whether you process a traffic for them or cover some of the function there. So that we do believe service provider is the most important part. We also do believe the service provider will come back up as the top sector for us. That's been like a high 20 like a few years ago, like 27%, 28%, 29%. I think last quarter, it's about 15%, 16%. So we do believe in the next few years, service provider will be go back on top, which including the SASE also leverage the 5G SD-WAN and some other security service they may offer. Like there's some key component. That's also the reason we're the first one to integrate into the OS level. We feel that's very, very important of making service provider more easy to manage. It's a SASE service offering and sometimes even beyond all this SASE. Basically, anything the service provider can help in the enterprise customer or customer even like a retail work from home to offer the security can be ramped up within the SASE. But on the other side, there's a more broad infrastructure need to be covered, whether the IoT/OT, work from home, connected car. There's a lot of other part also we kind of cover right now. It's not on the definition SASE, but we also see there's huge potential going forward.

So when you talk about service provider offering SASE, is it offering out to their customers so they're providing a managed service or selling the offering? Or is it utilizing it for their own benefit in their own network?

I think that most it's come from offer to the customer, just like how -- whether [ he's got ] some other one. But because service provider own most of the infrastructure themselves, they also have a very big customer base. So sometimes they see these carriers as their competitor. But somehow, some service providers are just a little bit behind on certain technology product investment, the service investment there, which we are helping them to catch up. And at the same time, offer better service, secure service to their customer base, which they have -- been there for a long time. And also because they own the infrastructure, is more easy and probably will be still more profit for them to offer the SASE compared to some SASE provider. They have to invest in some of the infrastructure and also still kind of -- much less coverage compared to the carrier like AT&T, BT, Orange, all these, they offer the same SASE, which they have a much bigger infrastructure behind to supporting them.

Yes. No, that makes sense. So I think one of those partnerships, I think, is AT&T that you announced back in March. How long does it take a partner like that to ramp to -- and what are the kind of the key milestones on what they need to do to get ramped?

I see the carrier service provider, not just this new service, but also any other product, they take usually about 1 to 2 years. But once they ramp up, they also last very long and the business can be pretty big, pretty huge. That's where we see the marketing come from AT&T and other. We feel pretty excited. We also see pretty healthy pipeline, pretty good growth from there.

Got it. Got it. Let's switch over to 5G. Devices are starting to show up in hands like my iPhone 5G-enabled, et cetera. But you kind of touched upon it, but what are the opportunities for Fortinet in 5G?

I feel the initial may come from like the IoT/OT side. The 5G, probably the first time they connect more devices, let's say, probably 10x more device compared to people that connect -- additional people get connected. So that's where we see the first benefit, is more come from the IoT/OT side of the 5G. So we have been working with a lot of carrier service provider, a lot of like the health care industry, some other manufacturer to see how to secure their kind of the 5G infrastructure. And eventually will go to the consumer. But right now, probably the OT -- IoT business side what be benefit first.

And how would you secure that? Is like -- is it remote access? Is it like a VPN for device? Or what would you actually sell to secure a 5G IoT implementation?

Because once they get connected, leverage whatever 5G, they need to have a better visibility and also need to monitor all the data, make sure it's all secure, try to avoid this, Colonial Pipeline kind of issue happening. And at the same time, it's -- 5G have to deal with much faster, bigger infrastructure. All need to be secured. Some work with service provider carriers, some work with sort of enterprise directly. But it's -- because a lot of them relate to the critical infrastructure, they really have -- whether the health care or some other, has to be -- there are certain regulations, certain standard have to secure all this data.

So on that, is that actual opportunity? It sounds like that's more inside the network operations side or the actual networks of the carriers supporting the security needs of that higher throughput bandwidth.

I think it's both. Definitely, the infrastructure also needs to be keep improving. That's where we have the product rather in a rough, in a ruggedized environment, and also there's some other part kind of more go to the data center, go to the NOC/SOC side to help them get better visibility to manage all this. But it's -- yes, that's also making the carrier service provider will be -- invest more or kind of involve more in this space. That's also one of the top reason we see they come back to be the #1, the top, for our business in the sector side.

So when you expanded out, so we talked SD-WAN, talked 5G. But when you kind of holistically look at what's driving growth in product in particular, because we know that drags along and gives you the long tail of the services revenue, what else is really driving the demand for that core part, just appliance-based product business?

I think still similar like in the last 20, 30 years, you're also probably the longest analyst to cover this. If you're getting more connectivity and there's more attack surface, then going forward, we don't see that part slow down. But so far, the networking, whether the 5G, they only talk about connectivity and speed. So anything on the content, on application, on the device, on the user, on the location has to be covered by security. That's also become more and more important. And also a lot of more valuable data, a lot of all things -- once you get connected, you also need to be kind of securely managed. So we just don't see any slowdown in the network security. But also for us, we also want to -- because security also, they need more computing power to process the same traffic compared to the networking, routing, switching. So that's where the AC has come up to that advantage, give additional computing power, also lower the cost. And at the same time, the other part is really, there's just so many different products to cover different part of infrastructure. So the cost -- the management cost is still very, very high for security. So how to help in consolidate? How to make the management more easy? So thus, you see some vendors, some product consolidation going on. So now it's more about kind of how to secure the whole infrastructure instead of just within the network security itself keeping the new generation. So that's where we see secure the whole infrastructure, cover most attacking surface. And long-term wise, like whether the edge computing, the immersive technology also need to be keep in mind to develop. That's where like Fortinet today, we also told the team it is very important to keep up the internal innovation and follow the chain quickly. That's also one of the reason this space gets so fragmented is because once some company get bigger, they all depend on acquisition for growth, to follow the chain, to get a new product. So if we can keep up the internal innovation, so we can keep inorganic growth for long term. And that tend to be better, more profitable and also working on the integration, on automation much better.

Well, also, it eliminates the risk of tech debt, right? So I remember all the way back to the beginning, the way that you built the platform so that you could plug in all the modular since you have the ASIC to get those hooks down to the chip level, and you can't just buy things and have that work because I think it would just eventually break.

Yes, exactly. And also, eventually, this chip also can go to like whether certain connected cars, some OT, some other parts, which -- that's also we do believe the future of the edge, probably will be more important compared to the cloud, which also the edge computing and some other part as also -- actually, we see more and more important part of infrastructure.

Could you actually see a point where you actually resell just the ASIC for something like the connected car, like the Cavium did for just security processors?

I think we are open for that. But right now, we're just so busy using the chip ourselves right now. But I think going forward, once the computing requirement, whether in the cloud and edge, become more and more important because ASIC, they can lower the computing costs by like 10x and also much better reliable. Just like how the TPU, how some of the GPU being deployed in the cloud or can be deployed in the field or in the edge, in all different edge, we do see that that's a huge potential, we can grow going forward.

Yes. That makes sense. Are you seeing any of your customers actually coming to you and saying, geez, you know what, we pushed off some infrastructure spend during the pandemic. Now that we're kind of returning to office. Is there any kind of snapback where you have to catch up for some of that infrastructure spend?

Maybe Keith can cover this better. We do see very healthy spending environment and going forward, at least in the next 1 to 2 years and better from own kind of pipeline, the few feedback. It's a pretty healthy environment. And yes, maybe Keith know also quite well.

Yes. It was -- 2020 was certainly an interesting year with the different set of puts and takes than maybe you would typically see. On the plus side, we saw organizations setting up work from home and standing up, study from home. On the other side, we certainly saw that competitive displacements, if you will, and the opportunity to save customers or new customers' money, I don't think that was high on the priority list throughout 2020. I think the CIOs and CISOs had -- easily understand more important things to focus on. Now as you shifted into 2021, and Ken kind of talked about it, the pipeline. I think there's another set of puts and takes, if you will, that are coming into play. So I don't know that people really sweat the assets, so to speak, in 2020. Because if they had -- might have seen something different in my renewal versus refresh rates, and I didn't really see that. But as you get into 2021, you start with the economic backdrop, you start -- then you work through just the talking points of what's happening with the volume of hacks and the publicity it's getting, whether it's SolarWinds or the dramatic increase in ransomware, and now we get this concept of ransomware-as-a-service. So I think the tailwinds, if you will, are certainly very different this year. And I don't know that we're really seeing that -- were there some things that were delayed last year? Sure, absolutely. Are we getting more opportunities to go compete for new customers that we had in 2020? Absolutely. But when we look at all those things combined, I think that's where -- when you hear Ken talk about the growth in the pipeline that we're seeing, I think that's what you're seeing in terms of the lift to the business.

Yes. That's the double the benefit like one since really the pandemic slowdown spending is starting to catch up, more like 2009. After the 2008 crisis, 2009, 2010, you can see the spend on infrastructure. On the other side, the security landscape, you can see that whether Sunburst or the exchange of the Colonial Pipeline, that's more like a 2012, 2013 after the Sony and Target case, that's what the security spending also benefit from that. So you get a double benefit, I think, in the next few quarters. So we -- that's probably the time we leave our boat in the cybersecurity space.

Yes. I think that's probably a good segue. I was going to say it's the end, but I think part of that tailwind is also Biden administration executive order around cybersecurity. As you look at that and look at the spending that's come behind it, how do you feel that benefits Fortinet moving forward?

Yes. The investment into the infrastructure, definitely, we see that cybersecurity is a very important part. We'll benefit from that. But on the other side, the carrier service provider, they're spending also what we see kind of a ramp up pretty quickly. So that's where -- so far, that's the decision we made a couple of quarters ago, invest in the growth. So we do see some of the benefit of that.

Yes. Do you think -- to that point, do you think the government spend specifically shows up quick in 2021? Or is it more of a 2022 before we see the dollars actually hit the results for vendors such as yourself?

They probably showed a little bit quicker than the carrier service provider, put in this way. That's also last quarter, I say, the government sector become #1 passing the service provider. Probably like a few years ago. Service providers the #1 carrier. Now the government spending in the last few quarters become the #1. But I see long-term wise, definitely service provider come back up.

But again, just to make sure for investors clear, when you talk about government, that's global government, state, local, everything, not just federal, correct?

Yes, exactly.

Yes. That makes sense. Keith, let's talk a little bit about the balance between the growth and the profitability is. As Ken mentioned, you decided to kind of step on the gas in terms of investment, but how are you kind of managing that balance?

Yes. I think we've been fairly open in terms of coming into the year with our framework that we've talked about during the Analyst Days, targeting that 25% operating margin. It can go higher if growth is slower. But we want to get -- stay close to that 25% average for operating margin. We came into this year and we saw the tailwinds. We knew what was happening in terms of the security industry. We saw the evidence of what was happening in the attack surface, the attacks that were happening in the conversations, the GDP, the stimulus, et cetera. So we felt that this was a year that tilts very clearly towards growth. Obviously, I think you saw that in the first quarter, and Ken talked about those numbers with 27% billings growth. I think that -- and we're coming into a second quarter, that's probably our easiest compare. That was the low watermark for us last year with COVID, and it's been a steady increase ever since. When we look at the totality, and we said, this is a year in which there's a significant opportunity. We check with the pipeline. We check with the sales capacity. And I think we're in a pretty good spot now as we came through the first quarter, as we look through the second half of the year.

Sounds good. And where is those incremental dollars in investment go? In other words, how much of that goes into headcount versus things like marketing programs?

Yes. And I kind of touched upon it a little bit. There is the art of keeping your sales capacity balanced with your pipeline, right? Ken is not shy about adding sales capacity. Let's be clear about that. Now on the other side, it's a matter of making sure we're making the investments in the marketing programs and things like branding like you're probably seeing behind us. I think that this is now an opportunity where Fortinet as a company has reached the maturity that they can leverage or perhaps even exploit that branding, maybe better than they could have when we were a smaller company. But all that comes into that, keeping that balance of watching the pipeline grow and keeping the sales capacity move in tandem with it.

Now does that mean I go and I look at the price pool and I add it up and I make sure that's in my sales and marketing expense for the quarter?

[indiscernible] handicap.

Keith, you reported 27% growth in billings in the quarter, but the guidance for the year is more kind of around 20%. Can you kind of just help investors understand what did you factor into that guide?

Yes. I think we're -- obviously, you saw us beat and raise in the first quarter, and we raised the full year. And you know the story line, right? We raised more than the beat. So you get a kind of sense of where our mindset is. And again, the guidance setting process really starts with those 2 inputs. I talk about pipeline, how you slice and dice it, and sales capacity. We did spend a little more time looking at supply chain. That was something that probably -- in a different way than we did in 2020 and the early part of it. Trying to understand our contract manufacturers commitments, trying to understand how the chip shortage could impact it. At the end of the day, we're somewhat blessed because we do maintain high levels of inventory. Our turns are right around 2, so 6 months of inventory on hand. But still that presumes in your sales forecast, and your manufacturing forecasts are completely in sync and you're aligned. So we looked at that a little bit. And I think that was part of the conversation we had internally in terms of where we ended up for the guidance for the quarter and for the year.

Yes, also because we invest early into the facility, the infrastructure, like all the warehouse and manufacturer, we see we have a less impact. We are in a much better position compared to some of our competitor during whether certain supply chain issue or certain shortage -- chip shortage. So because we have a -- like I say, we -- the inventory we invest there, the facility and the warehouse, even sort of manufacture we invest, give us much better position to go through this kind of shortage process.

Well, that makes a lot of sense. Let's talk a little bit geographically. I think international had a little bit more strength than here in North America in the quarter. Is that just based on where your investments have been? Or how should we see the geographic contribution playing out relative to the guidance you've given?

I think there's some related to the pandemic, how things have been evolving. There's also some part because in the past, we also consider what's the return of investment or what's the contribution margin because some international, we more leverage the channel, so they can cover a little bit more on the marketing, on the supporting side compared in the U.S., probably the vendor need to do direct marketing, the direct sales coverage. Like Keith said, once we get bigger and we can more leverage the branding, more easy to get into some big enterprise in the U.S. So that's why we do see the management starting shifting back towards the U.S., towards some big enterprise right now. And -- but it's both come from the some feedback, either from the recovery from the pandemic and also come from whatever the size and the potential going forward, we do see the invest more tier towards the U.S. right now.

Yes. I think Ken is spot on it. I would also add to that. I think the -- this notion of being the incumbent versus being the challenger, which kind of plays to our geographic mix, I think internationally, we do well as the incumbent in having a dominant market share in many, many countries. And during the pandemic, I think that was certainly an advantage. In the U.S., particularly in the large enterprise space that we're moving towards and moving into, less so, more of a challenger. And I think in 2020, going back to CIOs and CISO priorities, trying to dislodge the incumbent was more challenging. And why that's relevant is as we, as a country, now exit the pandemic and you're starting to see the economic recovery in all kinds of different numbers, I would say the expectations of our U.S. sales team are much, much higher now than they have been in the last several quarters.

Makes sense. And I definitely want to shift and talk a little bit about cloud security from this perspective. We talked about SD-WAN and SASE. But when you think about securing actual cloud workloads, I think there's confusion on the part of investors as to what the different components that are going to be necessary. I'm just kind of curious, when you look at it, what are the pieces of the cloud security pie that you want Fortinet to lead?

I think we involve in all part of our cloud security. But we also -- also we just not have the cloud only, we also have a lot of other part of edge, other part of [ infrastructure we're wanting to ] cover. And so far, we do see the healthy growth in the cloud security side, but we also see a lot of other potential. Sometimes we also want to play our advantage, right, like whether the ASIC has some other, which lower the computing -- secure computing power -- secure computing costs quite a lot, both in the cloud also and applies on the edge side. So that's where we just want to cover all parts of our infrastructure, all edge, not just the cloud. And the cloud, I also agree, it's a very important part. But also looking at the total addressable market like in 2025, the cloud is about $10 billion, but network security is still about $40 billion. And there's other part of infrastructure, also quite important. So we want to cover all of that.

So would it be fair to say that looking at it, the opportunity is securing the service provider and what the service provider is offering and maybe the cloud provider themselves. So maybe it's not going to be something that a company buys in an AWS marketplace, but underneath the covers is in that infrastructure is where the opportunity lies for you?

Yes. We found out that tend to be works better. It's like whether it's cloud provider or service provider, carrier, more as a partner instead of a more competing with them to get some customers. So we would rather partner with all this cloud provider, carrier, service provider.

And then last area is just, Keith, you mentioned kind of the -- in terms of the enterprise shift, what has been that mix enterprise versus some of the other segments? And where directionally do you think it gets to?

Yes. I think we've given some numbers on the Analyst Day that if I kind of round them and I figure out how to allocate the service provider, it's roughly 1/3, 1/3, 1/3. And again, it is biased a little bit -- mix shifts a little bit by geography. And so I think the real opportunity that we're very excited about is continuing to work and expand into the U.S. enterprise part of the market. The competitive advantage that we have, Ken has talked about it with the ASIC chip, the performance, the cost of performance, the integrated platform. I think we're -- we understand the challenge in front of us, and I think we have a pretty good sense of how we need to execute against that.

Yes. That makes sense. And geographically, outside the U.S., do you think just as a lagging indicator. So will the other regions kind of follow suit with what the U.S. is doing currently just over a longer period of time?

Well, certainly depending on the technology. We saw that with SD-WAN, for example. It took off fast in the U.S. and it followed elsewhere. But there are other parts of the business. OT, for example, actually went very, very fast in Europe and faster than it did in the U.S. And now you're Seeing the U.S. trying to get there -- not trying to, but catching up in that area. So as much as I came into with the bias thinking that the U.S. was always going to lead, it hasn't been the case. Sometimes, it's other geographies.

And then last question, we'll let you guys go in our final minute here. How do we think about capital allocation? You talked about the preference for internal development, but how do you think about share repurchase versus acquisition, et cetera?

Go ahead, Keith.

Look, I think the 3 places that we can invest readily. One is M&A, which is the traditional one. Another is returning share or cash to the shareholders through the buyback. And then we do -- because we do have a long-term focus, the ROI from real estate in various different forms, particularly where we have warehouses or concentrations of engineers, all make sense. On the M&A side, we're very committed to the single platform that's fully integrated, the common operating system that runs all the products. And while we have not historically done large integrations, one of the reasons for that -- or acquisitions, one of the reasons for that is the concern around integrations. They're hard. I mean when they're bigger, they get harder. And I don't think I'm here announcing one way or the other what our plans are in the future, but you can kind of see how the strategy works, if you will. In terms of buyback, I think we've talked before that we saw an opportunity in the market, more specifically Ken did last March and April, and we went into the market in 2020, fairly large, about $1 billion. And so I think that opportunistic nature about buybacks is something that we remain very, very comfortable with. And again, as long-term investors, with Ken and Michael particularly, the opportunities that come to us for real estate and various opportunities is something that we'll look at.

And stock picking, timing like that, Ken, you're going to put me out of the job.

Well, I think it's a lot of -- long-term investment is very important, whether in R&D, in infrastructure, in the team. So that's where we will -- we want to be positioned better in the next 5 to 10 years.

Makes perfect sense. Ken, Keith, thank you so much for joining us. We really appreciate it.

Thank you.

Thanks, Sterling.